The EU’s commitment to combatting violence against women: rhetoric or reality?

Catherine Briddick, Martin James Departmental Lecturer in Gender and Forced Migration at the Refugee Studies Centre of the University of Oxford - @CateBriddick

Background

The EU has, at its heart, a legal commitment to combat discrimination, including that based on sex, and to promote gender equality. It has however, been subject to sustained and justified criticism for its failure(s) to live up to these commitments, particularly in relation to its treatment of migrant and refugee women. The announcement by the Commission in 2016 that the EU would sign and conclude (ratify) the Council of Europe Convention on Preventing and Combatting Violence Against Women (the Istanbul Convention) was, therefore, warmly received by activists and academics alike.

 

The Istanbul Convention, for readers unfamiliar with it, is only the second international, legal instrument to focus on violence against women and the role of that violence in maintaining women’s inequality. The purposes of the Convention (set out in Article 2) are to protect women from all forms of violence and to prevent, prosecute and even eliminate violence against women and domestic violence. To achieve these purposes the Convention imposes on Parties a comprehensive range of obligations including that they:

-          adopt integrated, co-ordinated and properly resourced policies and programmes to challenge gender inequality, monitor and respond to violence against women (Istanbul Convention, Chapter II);

-          prevent violence through education, training and awareness-raising (Chapter III);

-          protect and support victims via a range of non-legal and legal measures (Chapters IV, V and VI);

-          investigate, prosecute and punish offenders (Chapter VI);

-          grant autonomous and/or renewable residence permits to migrant women who are victims of violence (Article 59); and,

-          ensure that refugee women’s claims for protection and dealt with in a gender-sensitive way (Arts 60 and 61).

Significantly, the rights and protections the Convention provides victims are to be secured by Parties without discrimination on any ground, including migration or citizenship status (Article 4(3)).

In force since 2014, the Convention has, at the time of writing, been ratified by thirty-three States (including Germany, Austria, Denmark, France, Italy, Spain, Sweden and Turkey) and signed by many others, including the UK.

The EU’s current approach

The Istanbul Convention itself envisages EU accession (Article 75), something that the EU can do to the full extent of its competences, as this blog has already discussed. The procedure to be followed involves the Council, following a Commission proposal and the consent of the European Parliament, adopting a decision which concludes the agreement. The agreement must identify the legal bases for the EU’s accession, bases derived from the EU’s legal competence (its ability or power) to act in a particular field. Once ratified, this agreement is binding on the institutions of the EU and EU Member States, to the extent that the EU has concluded the treaty. You can read more about this process here.

The EU’s competence in relation to violence against women is extremely broad. The legal bases under the Treaty on the Functioning of the European Union (TFEU) identified by the Commission in its Proposal for the Council on the conclusion of the Istanbul Convention were:

Article 16 (data protection), Article 19(1) (sex discrimination), Article 23 (consular protection for citizens of another Member State), Articles 18, 21, 46, 50 (free movement of citizens, free movement of workers and freedom of establishment), Article 78 (asylum and subsidiary and temporary protection), Article 79 (immigration), Article 81 (judicial cooperation in civil matters), Article 82 (judicial cooperation in criminal matters), Article 83 (definition of EU-wide criminal offences and sanctions for particularly serious crimes with a cross-border dimension), Article 84 (non-harmonising measures for crime prevention), and Article 157 (equal opportunities and equal treatment of men and women in areas of employment and occupation).

The Commission argued that it was appropriate to base a Council Decision signing the Istanbul Convention on Articles 82(2) and 84 TFEU because the ‘predominant purpose’ of the Convention is to prevent crime and protect victims. Selecting these bases would enable the EU to ‘exercise its competences over the entirety of the Convention’. Accordingly, the Commission’s draft Council Decision refers to Article 82(2) and Article 84 TFEU and refers to the EU signing up to the Convention as a whole.

The Council, however, took a radically different approach to that proposed by the Commission, taking not one but two decisions to sign the Convention in May 2017.

The first decision refers to Article 82(2) and Article 84 TFEU but states in Article 1 that:

The signing, on behalf of the European Union, of the Council of Europe Convention on preventing and combating violence against women and domestic violence with regard to matters related to judicial cooperation in criminal matters is hereby authorised, subject to the conclusion of the said Convention (emphasis added).

The second decision identifies Article 78(2) TFEU (on the establishment of a Common European Asylum System) as its legal base, stating in its Article 1 that the signing of the Convention is ‘with regard to asylum and non-refoulement’ (again, emphasis added).

These decisions limit the legal obligations the EU will accept in relation to the Istanbul Convention only to those that concern judicial cooperation in criminal matters and to asylum and non-refoulement (not, for example, European Union free movement law).

These decisions not only diverge from the Commission’s proposal, but also from the EU’s position in relation to the UN Convention on the Rights of Persons with Disabilities (the CRPD). The Council decision which signed the CRPD signed the Convention as a whole and took as its legal basis the EU’s commitment to non-discrimination.

(Update: in April 2019, the European Parliament decided to ask the Court of Justice about legal issues relating to the EU's proposed conclusion of the Istanbul Convention).

Commentary

Readers of this blog can be forgiven for asking if any of the above really matters. The obligations that the EU is planning to assume under the Istanbul Convention are significant, even if they are more limited than many hoped for and anticipated. EU action in either of the two areas it has signed up to could yield significant improvements in the way that violence against women is responded to, at both an EU and national level.  

Well it does matter, for at least two reasons.

First, EU free movement law disadvantages women, including women who have been subject to violence. To take just one example, as this blog and I have argued, the CJEU’s shameful decision in NA (which concerned Article 13 of the Citizens Directive), left a third-country national woman whose EU citizen husband subjected her to domestic violence and then left the UK, without a secure migration status. Article 59 of the Istanbul Convention requires Parties to grant autonomous and/or renewable residence permits to victims of violence in a broader range of circumstances than that currently provided for by EU law, potentially improving the position of women like NA whose migration status is (or was) dependent on their partner. The EU’s decision not to sign up to this provision means that women who are subject to violence whose migration status is determined by EU law will continue to face considerable hardship.

Second, the Convention itself and the holistic approach it adopts to violence and discrimination against women have been attacked and undermined by States who are either Parties or signatories to it.  Some States, including Poland, Latvia, Lithuania and Croatia have sought to limit the obligations the Convention imposes by making impermissible and potentially invalid declarations / reservations to it. In Bulgaria the process of ratifying the Istanbul Convention has been halted following a controversial judgement from its Constitutional Court that the Convention contradicts Bulgaria’s constitutional protection of women as mothers. The Commission has expressed concern about these developments and has sought to encourage States to ratify the Convention fully and without delay, highlighting its own role as a potential enforcer of the Convention where EU competences are involved. The EU’s ability to provide either political leadership or legal support on these issues is, however, hampered by its own partial and highly selective engagement with the Convention. Not only is the EU open to allegations of hypocrisy, but its own actions give succour to, rather than challenge, the conduct of the very States whose behaviour it seeks to influence.

Can the Council’s position be challenged?

Whether the Council’s two decisions will actually lead to the EU engaging with the Istanbul Convention in the very limited way described here is, as yet, far from clear.  

The European Parliament has stated that it ‘regrets’ the Council’s approach because it raises ‘legal uncertainties as to the scope of the EU’s accession, as well as concerns regarding the implementation of the Convention’. The Parliament has recommended instead ‘a broad EU accession to the Convention without any limitations’. It is not known whether the Parliament’s ‘regret’ will extent to withholding its agreement to the Council’s decisions.

The decisions could also be subject to legal challenge. The Commission has successfully challenged a Council decision to enter into a legal agreement with a third country on the grounds that it did not proceed on the correct legal bases. A similar challenge, based on the arguments the Commission advanced in its Proposal (as discussed briefly above) or, more persuasively in my view, on the grounds that the EU’s legal response to violence against women should be based in its commitment to combat sex discrimination, may well yield success.

The Istanbul Convention is monitored and enforced by a committee, the Group of Experts on Action against Violence against women and Domestic Violence (GREVIO) via a reporting and inquiry procedure. The EU’s approach to the Istanbul Convention could also be challenged by GREVIO or another Party to it. Article 75 of the Istanbul Convention refers to the Convention as a whole being open for signature, not parts of it whilst Article 73 provides for a dispute mechanism to be created if Parties disagree over the application of its provisions.

At the time of writing the EU’s (stalled) ratification of the Istanbul Convention is being considered by the Council working party on Fundamental Rights, Citizens’ Rights and Free Movement of Persons (FREMP). This scrutiny is being accompanied by a concerted, EU-wide campaign in support of the Istanbul Convention and the EU’s full ratification of it. Over the next few months we will be able to gauge the impact of these processes, as the EU proceeds (slowly) to conclude the Convention. We will then find out whether the EU’s rhetoric on violence against women is any more than that. 

Barnard & Peers: chapter 9, chapter 20

JHA4: chapter I:5

Photo credit: Council of Europe

What do the 'no deal' notices tell us about Brexit and health?

Sarah McCloskey and Professor Tamara Hervey, University of Sheffield

The clock ticks towards Brexit Day and time to devise an agreement dwindles. The odds on reaching one are said to be at 50:50. This means increased incentives for both the UK and the EU to set out exactly what the implications of the UK leaving the EU without a legally binding Withdrawal Agreement (‘no deal Brexit’) would be. The EU began doing this in January 2018, through a series of ‘Brexit preparedness notices’. The UK issued its equivalents in August.

Here we focus on the UK notices, with reference to their EU counterparts, considering their implications for the health sector.

Overall, the UK’s notices attempt to pursue two conflicting functions. This leads to a pervading sense of contradiction. On the one hand, they aim to assure and provide certainty. Simultaneously, however, they set out to paint a sufficiently bleak picture to seek to deter negotiators from such an outcome.

Indeed, contrary to the idea propagated by some ministers that defaulting to WTO rules would be favourable, each notice begins with a statement that the statements merely represent a responsible government prudently preparing for all eventualities in the context of otherwise promising negotiations.

The Secretary of State for Health and Social Care has taken this one step further by advising that the work already done gives a clear basis for the health sector “to plan so that patients can continue to receive high-quality care unhindered.”

Said planned actions outlined within the respective notices include:

-       Stockpiling a six-week supply of medicines to allow for delays as a result of e.g. border controls.

-       Setting up new national systems - e.g. a national portal to which regulatory information about human medicines and devices can be submitted directly.

-       Increasing the role of the Medicines and Healthcare products Regulatory Agency (MHRA), including in areas such as batch testing medicines and approval of paediatric and orphan medicines. This is to account for the cessation of European Medicine Agency (EMA) membership. Its details of this are to be subject to an autumn consultation.

These actions attempt to counter the systemic and regulatory black holes that would result from what the EU has described as a ‘cliff-edge scenario’.

Regulatory alignment - the solution in the EU (Withdrawal) Act 2018 - does not remove the need for contingency planning. For example, despite the UK’s plan to retain regulations on blood and blood product standards, to import into the UK, it will be necessary to show compliance with UK standards, which are to be a continuation of EU standards. But the systems by which such compliance is shown involve cooperation and mutual recognition, not unilateral behaviour.  Similarly, the UK government’s intentions to align with the 2019 Clinical Trials Regulation ‘wherever possible’ when it comes into force post-Brexit Day won’t prevent the UK’s exclusion from the new EU clinical trials database. (Indeed, the EU has advised that efforts have already begun to remove the UK from several existing databases once it exits the EU.) Every example where there are elements of reciprocity in EU-UK relations is similar. UK patients will be excluded from the EHIC scheme. The system of sharing information about medical professionals won’t apply any more, so the UK won’t be notified of concerns about EU-27 doctors working in the UK.

On a no-deal Brexit, the UK will be immediately excluded from EU decision-making processes. The UK will no longer be represented at committees which decide on safety of certain medical devices (including cardiac pacemakers and breast implants). It has already been excluded from some EMA processes, on the basis that the person chairing the EMA committee has to be present for the whole period of decision-making, and that’s not the case for the MHRA.

One of the EU’s strengths is in its collective decision-making powers, used to devise common regulations. Of course, no regulatory system is perfect, but on leaving the EU, the UK loses many of the trade-improving, burden-easing, and practice-refining benefits that come with economies of scale.

And, for the health sector, there are no clear counter-advantages. The EU regulations are a product of the member states’ views; the UK’s voice has been heard and expert research has been considered to create them. Thus, as highlighted by the UK’s notices, there is no clear motivation to do anything other than copy and paste them. But, as the EU has stated unambiguously, the UK’s fundamental change in status to a third country means that this will by no means equate to continuity.

Barnard & Peers: chapter 27, chapter 21

Photo credit: Royal College of Nursing

Analysis of the ECtHR judgment in Big Brother Watch: part 1

Lorna Woods, Professor of Internet Law, University of Essex

This chamber judgment is the latest in a line of cases that deal with secret surveillance, a topic which seems to be appearing increasingly frequently in a post-Snowden world. This judgment is substantial (over 200 pages in length) and deals with three cases challenging the UK’s now mainly repealed Regulation of Investigatory Powers Act 2000 (RIPA) as regards to interception of communications in bulk, the acquisition of communications data and the sharing of intercepted communications and communications data between the UK and the United States of America: Big Brother Watch (app no. 58170/13), Bureau of Investigative Journalism and Alice Ross (app no. 62322/14) and 10 Human Rights Organisations (app. no. 24960/15).  It follows in the steps of the Liberty case (app no. 58243/00) against the previous regime and, given the similarity between some aspects of RIPA and the Investigatory Powers Act 2016 (IPA), might have relevance for our understanding of that act too. In addition to questions about Article 8, the judgment also deals with the impact of surveillance on freedom of speech under Article 10 ECHR. 

This post is the first of two on the judgment. It outlines the issues and the Court’s reasoning. The second comments on the judgment. Given the size of the judgment that will be just an initial reaction to the judgment – there will, no doubt, be much more to be said.

Factual Background

The applicants in the three cases are organisations and individuals who are either journalists or are active in campaigning on civil liberties issues. Their challenges to RIPA were triggered by the information revealed by Edward Snowden which made apparent the existence of surveillance and intelligence sharing programmes operated by the intelligence services of the United States and the United Kingdom.  Specifically, they believed that the nature of their activities meant that their electronic communications and/or communications data were likely to have been intercepted or obtained by the UK intelligence services relying on the regime found in RIPA.  Three areas of problems were highlighted:

-          bulk interception of ‘external’ communications under s. 8(4), as well as connected communications data;

-          the sharing process whereby the British agencies received data collected by the US; and

-          access to communications data under Part II RIPA.

In all instances the applicants thought that the protection against abuse were insufficient and that the regimes were neither lawful nor necessary in a democratic society.

Only the applicants in the third case brought an action before the Investigatory Powers Tribunal (IPT), alleging violations of Articles 8, 10 and 14 of the Convention.  Although the IPT found two ‘technical violations’ of the Convention, in the main it regarded the challenged regime to be in accordance with the requirements of Article 8, notably the requirements set down in Weber and Saravia (app no. 54934/00).

Judgment

The first issue concerned exhaustion of domestic remedies, in particular the failure to bring a case before the IPT.  The applicants argued that in the light of the ECtHR’s own ruling in Kennedy (app no. 26839/05), the IPT would not be an effective remedy and they were therefore not obliged so to do.  The Court agreed with this assessment of its case law in general terms, but now thought that recent practice showed that the IPT now constituted a viable route for a remedy, especially given the response of the UK government to its findings. Nonetheless, the Court accepted that, at the time the applicants in the first and second of the joined cases introduced their applications, they could not be faulted for having relied on Kennedy as authority for the proposition that the IPT was not an effective remedy for a complaint about the general Convention compliance of a surveillance regime. It therefore found that there existed special circumstances absolving those applicants from the requirement that they first bring their complaints to the IPT.

The Court first considered the position under s. 8(4) RIPA and whether it met the tests of legitimate purpose, lawful and necessary in a democratic society. In doing so, it noted that there was jurisprudence in this field but that in previous jurisprudence the Court had distinguished between different types of secret surveillance, finding that there different levels of intrusion depending on the data collected, and also different rules depending on whether national security was in issue.  The Court sought to synthesise the principles, suggesting that the 6 principles established in Weber – to ensure the lawfulness of any such regime - were the starting point, though they might need to be differently applied depending on the type of surveillance. These need not be updated to take account of changes in technology.  These minima are:

-          the nature of offences which might give rise to an interception order;

-          definition of the categories of people liable to have their communications intercepted;

-          a limit on the duration of interception;

-          the procedure to be followed for examining, using and storing the data obtained;

-          the precautions to be taken when communicating the data to other parties; and

-          the circumstances in which intercepted data may or must be erased or destroyed.

In the context of national security it also recognised the gloss added by the Grand Chamber in Zakharov (app no. 47143/06) the review mechanisms and remedies should also be taken into account. The Court noted that the nature of secret surveillance was such that until an individual were to be notified about such surveillance, that individual would not be in a position to exercise their rights. In this, the safeguards against abuse assumed high importance; moreover, the role or rights to remedies was important for protection after notification.

Looking at the situation in issue, the Court started by making the general point that operating a bulk interception scheme was not in itself in violation of the Convention. Governments would have “a wide margin of appreciation” in deciding what kind of surveillance scheme was necessary to protect national security.  The operation of the system would still however need to be checked to ensure that there were sufficient safeguards against abuse.  The applicants argued that the fact that there was no requirement for prior judicial authorization was a fatal flaw in the scheme. 

The Court agreed judicial authorisation was an important safeguard, perhaps even “best practice”, but by itself it was neither necessary nor sufficient to ensure compliance with Article 8. It was unnecessary because of the ex post controls available in the British system. Looking to Zakharov, the Court recognised that a formal requirement was insufficient – the requirement there had not prevented bad practice. The Court then held that regard had to be had to the actual operation of the system of interception, including the checks and balances on the exercise of power, and the existence or absence of any evidence of actual abuse.

In assessing the scheme the Court took the law at the time of its consideration of the claims; this meant that the Court considered the matter after the impact of the Snowden leak and some of the consequent changes to practice, including revisions to relevant codes accompanying RIPA, as well as statements in Parliament (such as the clarification as to what an external communication was – it includes Google searches, tweets and Facebook posts from by users in the UK). 

The Court took the view that, as regards the first Weber requirement, the law was clear as to the circumstances in which and the conditions on which a section 8(4) warrant might be issued. There was no evidence to suggest that the Secretary of State was authorising warrants without due and proper consideration. The authorisation procedure was subject to independent oversight and the IPT had extensive jurisdiction to examine any complaint of unlawful interception. Following its analysis in Kennedy, the Court accepted that the provisions on the duration and renewal of interception warrants, the provisions relating to the storing, accessing, examining and using intercepted data, the provisions on the procedure to be followed for communicating the intercepted data to other parties and the provisions on the erasure and destruction of intercept material provided adequate safeguards against abuse.

There were some weaknesses in the system.  While in the opinion of the Court the selectors (e.g. email address) and search criteria used to narrow down the mass of information collected to that which would be read by analysts did not need to be made public or be listed in the warrant ordering interception, the choice of search criteria and selectors should be subject to independent oversight (para 387); indeed the Court expressed some concerned about the cables (‘bearers’) selected for tapping. Here the ex post review by the Interception of Communications Commissioner (now replaced under the IPA by the Investigatory Powers Commissioner) and, should an application be made to it, the IPT were held not to be ‘sufficiently robust to provide adequate guarantees against abuse’ (para 347).

The Court also expressed concern about communications data.  This is often summarised as who, where, when but this underplays the significance of the data collected.  Indeed, here the Court rejected the Government’s argument that communications data was necessarily less sensitive than the content of the communications (para 357). The Court explained the position thus:

... the content of an electronic communication might be encrypted and, even if it were decrypted, might not reveal anything of note about the sender or recipient. The related communications data, on the other hand, could reveal the identities and geographic location of the sender and recipient and the equipment through which the communication was transmitted. In bulk, the degree of intrusion is magnified, since the patterns that will emerge could be capable of painting an intimate picture of a person through the mapping of social networks, location tracking, Internet browsing tracking, mapping of communication patterns, and insight into who a person interacted with. (para 356)

In the context of s 8(4), communications data associated with the communications intercepted is also covered by the warrant but crucially some of the limitations (e.g. that the communication must be external) do not apply to this data.  The Court concluded that the unjustified lower level of protection meant that there was a violation in this regard.

The Court then considered the data sharing arrangements, the first time that the Court had been asked to consider the matter. It noted to start with the many ways in which this issue might arise.  The interference in the case had not been occasioned by the interception of communications itself but lay in the receipt of the intercepted material and subsequent storage, examination and use by the intelligence services.  It confined its judgment to the specific argument brought before it: the breach occasioned by the British services receiving American intelligence.  The applicants argued that this indirect access should be treated the same way as direct surveillance by the British services.  The Court commented that:

"[a]s with any regime which provides for the acquisition of surveillance material, the regime for the obtaining of such material from foreign Governments must be 'in accordance with the law'..., it must be proportionate to the legitimate aim pursued, and there must exist adequate and effective safeguards against abuse .… In particular, the procedures for supervising the ordering and implementation of the measures in question must be such as to keep the 'interference' to what is 'necessary in a democratic society'" (para 422).

The Court also recognised the danger of States using intelligence sharing as a means to circumvent controls (para 423).  It nonetheless accepted that the safeguards need not look identical in this context as in that of direct surveillance. Applying the principles to the facts, the Court found unanimously that there had been no violation. In particular, it accepted that the lawful requirement had been satisfied although the basis for the data sharing was an internal agreement which were disclosed only during proceedings before the IPT and subsequently incorporated into the Interception of Communications Code (para 426). The Code links the circumstances in which intelligence may be requested to the issuing of s. 8(1) or s. 8(4) warrants, thus circumscribing the circumstances in which such requests may arise and indirectly imposes supervision via sign-off by the Secretary of State and review by the ISC and the Interception of Communications Commissioner. 

The Court applied its assessment of the Code’s safeguards in relation to s. 8(4) warrants (in paras 361-363) here. Its assessment of the proportionality of information sharing was influenced by the threat of international terrorism and the global nature of terror networks necessitating information flow. In the Court’s view, ‘this “information flow” was embedded into a legislative context providing considerable safeguards against abuse’ so that ‘the resulting interference was to that which was “necessary in a democratic society”’ (para 446) and it considered that the threshold set by the Venice Commission – that the material transferred should only be able to be searched if all material requirements of a national search were fulfilled – were met (para 447). 

The next issue was the final question relating to Article 8.  It concerned Chapter II of RIPA which allows specified authorities to access communications data held by communications service providers (CSPs).  As noted, communications data is not necessarily less intrusive than content.  The Court did not however go into detail on this here, although it noted that real time surveillance is more intrusive that the transfer of records of existing data (citing Ben Faiza (app no. 31446/12)). It re-iterated that the same three criteria apply: lawfulness, legitimate aim and necessary in a democratic society.  The Court focussed on the lawfulness of the rules, referring to the position under EU law – notably Digital Rights Ireland (Case C-293/12 and C-594/12) and Watson (Case C-698/15) - which requires that any regime permitting access to data retained by CSPs was only to be for the purpose of combating “serious crime”, and that such access be subject to prior review by a court or independent administrative body. RIPA -although it provided a clear basis for action on the face of it - did not comply with this requirement and was therefore was not compliant with domestic law requirements (para 467).

A further issue arose in the Bureau of Investigative Journalism (BIJ) complaint. There, BIJ (a newsgathering organisation) and a journalist (Ross) raised the issue of interference with confidential journalistic material occasioned by the operation of both the section 8(4) and the Chapter II regimes.  While the Court has emphasised the importance of protection of journalists’ sources, its case law has distinguished between court orders for disclosure and searches carried out by the authorities to obtain this information – that latter is more intrusive. Further, the Court also distinguished between attempts to reveal sources and investigations into the commission of crimes. So the importance of source confidentiality is not an automatic trump card. The Court noted that the s. 8(4) regime was not aimed at monitoring journalists or uncovering journalistic sources.  The authorities would often only know that a journalist’s communications had been intercepted when examining the intercepted communications. Following Weber, this in itself could not be characterised as a particularly serious interference with freedom of expression. Nonetheless, where those communications were selected, the concerns would increase and safeguards would be required, especially as regards the need to protect confidentiality. In this context, concerns expressed in relation to the s. 8(4) regime ran through to Article 10 concerns. The Court emphasised that:

... there are no [public] requirements...either circumscribing the intelligence services' power to search for confidential journalistic or other material (for example, by using a journalist's email address as a selector), or requiring analysts, in selecting material for examination, to give any particular consideration to whether such material is or may be involved. (para 493)

This blanket power without any “above the water” arrangements limiting the intelligence services’ ability to search and examine such material constituted a violation of Article 10.

As regards the Chapter II regime, while there were some protections in place for journalistic sources, the Court determined that this was limited. They applied only where the purpose of the application was to determine a source. They would not apply in every case where there was a request for the communications data of a journalist, or where such collateral intrusion was likely.  Given this and the fact that access was not limited to ‘serious crime’, the Court found a violation of Article 10.

The Court rejected complaints under Article 6 as well as Article 14 combined with Articles 8 and 10 of the Convention as manifestly ill-founded.

The judgment was not unanimous. Judge Koskelo, joined by Judge Turkovic, disagreed with some points of the reasoning of the majority and particularly the appropriateness of relying on old case law in a context following a technological ‘sea change’ in which people’s lives are more thoroughly exposed to view.  Judges Pardalos and Eicke did not agree that the applicants in the first and second case should have been absolved from the requirement to exhaust domestic remedies, nor – in the light of the recent chamber judgment in Centrum For Rattvisa (app no. 35252/08) – that there had been a violation of Article 8 in relation to s. 8(4) warrants.

Barnard & Peers: chapter 9

Photo credit: Journalism, Media and Culture

Lock ‘em up: the proposal to amend the EU’s Returns Directive

Steve Peers, Professor of Law, University of Essex

One of the key EU laws regulating migration is the Returns Directive, now nearly ten years old, which regulates the main aspects of irregular migration by non-EU citizens. It requires Member States in principle to issue a return decision to every non-EU citizen not authorised to be on their territory, and to enforce that decision by removing the person concerned. In some cases, irregular migrants should be given a period for voluntary departure, so they won’t be subject to forced removal. Member States are required to issue entry bans to many of those subject to a return decision, but must provide for basic remedies to challenge expulsion. There are also rules on the grounds for detention of irregular migrants and their conditions during detention.

When it was adopted, NGOs supporting migrants believed that the Directive set an inexcusably low standard. However, in practice the CJEU has interpreted the Directive more liberally than some expected. I discussed this in detail in a paper published in 2015, and there have been several more CJEU judgments interpreting the Directive since then: Celaj (discussed here), Affum, Ouhrami, Gnandi and KA. For instance, the case law limits the use of the ordinary criminal law to jail irregular migrants, requiring Member States to use the administrative detention rules in the Directive in most cases. The jurisprudence also establishes a limited right to a hearing for irregular migrants (discussed here), and in some extreme cases provides for a basis to prevent expulsion of irregular migrants, and secure their basic rights to health care in the meantime (as discussed here).

The case law has been sufficiently liberal that in light of the perceived refugee ‘crisis’ of the last three years, it’s the EU Commission and Member States who are now concerned about the law, because in their view it is not stringent enough. This approach was set out in an ‘action plan’ and a recommendation issued by the Commission in 2017, as well as a revised Returns Handbook. They represented a change of tone from the 2014 Commission report on the implementation of the Directive, discussed here.

In 2014, the Commission urged Member States to apply the Directive generously, while in 2017 it recommended that they apply it as strictly as possible. But there are limits to the latter strategy without amending the legislation to lower standards, and today the Commission did just that, tabling a proposal to amend the Directive. To become law, it must still be agreed by the European Parliament and the Council (Member States’ interior ministers). It forms one of a number of measures proposed or adopted in recent years to strengthen migration control in light of the perceived crisis, including changes to the EU’s Frontex border agency (discussed here), and greater use of the EU’s Schengen Information System to enforce entry bans and expulsion orders (the agreed text is here).  Another proposal to give further powers to Frontex – including to assist more with expulsions – was also released today.

The UK, Ireland and Denmark opt out of this Directive – although it will apply to UK citizens in the participating Member States after Brexit, in the event that they are irregular migrants.

The new proposal

The first change in the proposal is to define ‘risk of absconding’. The effect of such a definition is to make it easier to refuse a prospect of voluntary departure, and correspondingly easier to justify detention. Reducing the number of people given the chance of voluntary departure will in turn increase the number of those subject to an entry ban.   

The list of cases which might constitute a risk of absconding is long (there are sixteen factors listed) and non-exhaustive (‘at least’). Some of the grounds are very broad (‘illegal entry’). Member States will have to apply these criteria on a case-by-case basis, but four of the grounds create a rebuttable presumption: using false documents et al; opposing expulsion violently or fraudulently; not complying with a measure like a reporting requirement; or violating an entry ban.

There’s a new obligation for irregular migrants to cooperate with the authorities, which reflects CJEU case law such as KA. The proposal does not, however, add the right to a hearing expressly into the Directive. Member States have a new obligation to issue a return decision as soon as a migrant loses a right to a legal stay, or an asylum seeker’s application is turned down at first instance. In theory this just repeats the underlying obligation to issue a return decision to all irregular migrants, but the Commission states that not all Member States do this. The issue of a return decision after a first-instance refusal of an asylum application takes account of the Court’s judgment in Gnandi: asylum-seekers cannot be regarded as irregular migrants until a refusal of their application at that stage. The implications of this for appeals are discussed further below.

Voluntary departure is tightened up. First of all, Member States need no longer give a seven-day minimum of time for an irregular migrant to depart. Secondly, the three cases where Member States can opt to refuse to give the irregular migrant a chance to leave voluntarily – risk of absconding, manifestly unfounded or fraudulent application for legal stay, and risk to public policy, public security and public health – are replaced by an obligation to refuse the chance of a voluntary departure in such cases. As noted above, the effect of this change is bolstered by including a wide definition of what might be considered as a risk of absconding. Although the CJEU has narrowly interpreted the exception for public policy, et al, as discussed here, this matters less now that the ‘risk of absconding’ ground is widely defined.

There’s a new obligation to try to confirm the identity of the irregular migrant who doesn’t have a travel document, and also to obtain such a document. The thinking is presumably that this should facilitate the expulsion process. Non-EU countries sometimes insist on such documentation before readmission, and as the Commission notes, its proposal is consistent with separate recent proposals to amend the EU’s visa laws (discussed here) to punish non-EU countries for non-cooperation.

Next, there’s a change to the rules on entry bans. Member States may impose an entry ban on an irregular migrant without making a return decision, if they detect the irregular migrant on his or her way out of the EU. This would not be an expulsion measure, but a means of trying to prevent the person concerned from coming back in future. Imposing an entry ban would only apply in such cases ‘where justified on the basis of the specific circumstances of the individual case and taking into account the principle of proportionality.‘

As noted above, there’s also an implied change to the rules on entry bans too. Since an entry ban must be imposed when an irregular migrant is refused the possibility of voluntary departure (subject to exceptions in individual cases), narrowing down the cases where voluntary departure is possible will have the knock-on effect that more entry bans are issued.  

Next, there’s a new obligation to set up a ‘return management system’, comparable to Member States’ obligations relating to border control and asylum. This will have no direct impact upon irregular migrants.

There are several changes to the rules on remedies. First, any remedy must be before a judicial authority, not an administrative authority. This implicitly takes account of recent CJEU case law on appeals against refusals to issue a visa (El-Hassani), where the Court said that the EU Charter of Rights requires judicial control of immigration decisions.

Next, a new clause states that failed asylum seekers have only one instance of appeal against a return decision, if they have already had effective judicial review within the asylum process. This transposes the recent Court judgment in Gnandi. A further new clause, also taking account of CJEU case law, states that where the irregular migrant has concerns about refoulement (being sent to an unsafe country), at last the first level of appeal must have suspensory effect, ie stopping removal from the country. The irregular migrant can ask for suspensory effect in the event of a further appeal, but the national court must rule on that request within 48 hours. Furthermore, these possibilities don’t exist where there have already been proceedings concerning asylum or legal migration status, unless there are new issues in the case. Finally, failed asylum seekers will have only five days to appeal a return decision.

The proposal then moves on to detention. There will now be three grounds for detaining irregular migrants, rather than two; and anyway this list will become non-exhaustive (the word ‘only’ will be deleted). The first ground – risk of absconding – will be broadly defined, as we have already seen. The second ground, which was already broadly defined, remains ‘avoids or hampers the preparation of return or the removal process’. The new ground is where the irregular migrant ‘poses a risk to public policy, public security or national security.’ This new ground matches one of the grounds to detain asylum seekers in EU asylum legislation, which the CJEU has interpreted narrowly (as discussed here); but this hardly matters when the other two grounds for detention are broadly defined, and the whole list is non-exhaustive.

Another change relates to time limits: Member States’ maximum time for detention must be at least three months. This will not mean that all irregular migrants must be detained for that long, only that this must be a possibility on the books as a maximum period of detention. The other current rules on detention time limits – six months as the normal time limit, a further twelve months as a possibility in special circumstances – are retained.

Finally, a new clause sets out special rules for failed asylum seekers at border posts, effectively derogating from some key standards in the Directive. They must be given a standard ‘tick-box’ form setting out the return decision, rather than a reasoned explanation. In principle there’s no chance for voluntary departure, except where the migrant holds a valid travel document (handed over to the authorities) and cooperates fully. Such migrants will have only 48 hours to appeal a return decision, and suspensive effect only applies where there are significant new findings or there was no effective judicial review already. Detention is apparently always justified, with a four-month time limit; but then the proposal provides for the regular time limits to be applied on top of this, if return is not possible.

Comments

This proposal is entirely concerned with facilitating the expulsion of irregular migrants, and detaining them to that end – in addition to imposing entry bans to make sure they do not return. The narrower possibilities to obtain a period of voluntary departure will mean surprise knocks at the door, detention time and forced removal for more irregular migrants. More legal challenges will be fast-tracked, with the time limits in this proposal arguably below the standards set by CJEU case law (see the Diouf judgment). More migrants will be detained, and the Member States with the most generous approach to detention time limits will have to be more stringent.

While the proposal does not directly affect the case law that limits Member States’ use of criminal law to detain irregular migrants, it will to some extent circumvent the limits in that case law indirectly, by giving Member States more powers to detain irregular migrants in the context of administrative law instead. Note, however, that the rules on detention conditions would not be changed; and the continued requirement to channel immigration detention largely away from the criminal law process should prevent the family separation we have seen in recent months in the USA, which results from applying criminal law detention to irregular migrants who are parents.

Will the proposal impact upon the declining rates of expulsion? (45% in 2016 and 36% in 2017, according to the Commission) It’s hard to tell, because there’s no proper impact assessment alongside the proposal: this is not evidence-based policy-making, or at least not transparently so. Some of the proposals might possibly backfire: for instance, if the consequence of an irregular migrant leaving the EU is being detected and then subjected to an entry ban, the irregular migrant concerned might just decide not to leave at all.

In any event, the effectiveness of expulsion policy depends partly upon cooperation of non-EU States, and this proposal can’t affect that – although, as noted above, there are other EU initiatives underway on that front. Detaining more people has a cost for national budgets, but this proposal overlooks this awkward fact. Anyway, without additional cooperation from outside the EU, constructing more detention centres will not by itself increase the rate of expulsion, but merely increase the cost of irregular migration for national budgets and the misery of the persons concerned.

Peers & Barnard: chapter 26

JHA4: chapter I:7

Photo credit: Robert Hickerson on Unsplash 

Religious discrimination at work: Can employees be fired for getting divorced?

Ronan McCrea, Professor of Constitutional and European Law at University College London

The Framework Directive on Discrimination in Employment came into force in 2003 but it took 14 years for the Court of Justice to issue its first major decision on its religion-related provisions. However, we have now had four major decisions in the last year and a half so the precise impact of the Directive’s religion-related provisions is now becoming clear.

We now know that a consistently-applied neutrality requirement that prevents employees wearing symbols of religion or belief at work will be considered to be indirectly, not directly discriminatory (the Achbita case), that the court takes a narrow view of what counts as a genuine and determining occupational requirement justifying direct discrimination on religious grounds (Bougnaoui and Egenberger cases) and that the ability of religious employers to restrict roles to co-religionists in order to protect their ethos was subject to a proportionality test (the Egenberger  case.) (On Achbita and Bougnaoui, see discussion here; on Egenberger, see previous discussion on this blog here)

The latest judgment, in the case of IR v JQ addressed the matter of discrimination on grounds of religion addressed by the Court in Egenberger, but also covered the final major outstanding issue in relation to the religion-related provisions of the Directive; the scope for those employers, termed organisations ‘the ethos of which is based on religion or belief’ by the Directive, to impose an obligation on their employees to behave with loyalty towards the religious ethos of their employer.

However, it did so in a slightly unusual way because in this case the employee in question contested the fact that he had been placed under a greater duty of loyalty to his employer’s ethos because he belonged to the same faith as his employers. In other words, he alleged that the obligation of loyalty had been applied to him in a discriminatory way rather than simply alleging that an excessive obligation had been imposed.

In this case the employee, ‘JQ’, was the head of internal medicine in a hospital run by ‘IR’. IR is a not-for-profit, Catholic organization that runs a number of organisations including hospitals as part of what the Court of Justice called ‘an expression of the life and nature of the Roman Catholic Church’. JQ is a Roman Catholic. He was married in a Roman Catholic ceremony but divorced in early 2008. He subsequently married a new partner in civil ceremony. When his employer became aware of this he was fired from his post in March 2009.

His former employer argued that the dismissal of JQ was justified because by remarrying, he had breached the duty (contained in his contract) to be loyal to the ethos of the Catholic Church. JQ argued that his dismissal amounted to impermissible discrimination on the basis that an employee who was not a Catholic would not have been fired for entering into a second marriage.

EU law does allow (but does not require) Member States maintain in force exemptions from the duty not to discriminate. Article 4(2) of the Directive states: 

‘…. in the case of occupational activities within churches and other public or private organisations the ethos of which is based on religion or belief, a difference of treatment based on a person's religion or belief shall not constitute discrimination where, by reason of the nature of these activities or of the context in which they are carried out, a person's religion or belief constitute a genuine, legitimate and justified occupational requirement, having regard to the organisation's ethos. This difference of treatment shall be implemented taking account of Member States' constitutional provisions and principles, as well as the general principles of Community law, and should not justify discrimination on another ground.’ 

  

The key issue in the reference was whether the scope of the exemption from the duty not to discriminate on grounds of religion or belief granted by German law to religious organisations was compatible with Directive 2000/78. JQ’s Catholic employers believed it necessary to place employees with managerial roles who shared their Catholic faith under a greater obligation of loyalty than that placed on non-Catholic employers. German law implementing Directive 2000/78 provides them with significant scope to do so. It provides that:

  

‘The prohibition of a difference of treatment on grounds of religion or belief shall not affect the right of the religious communities mentioned in subparagraph 1, institutions affiliated to them, regardless of their legal form, or associations that devote themselves to the communal nurture of a religion or belief, to require their employees to act in good faith and with loyalty in accordance with their self-perception.’(paragraph 9(2) of the Allgemeine Gleichbehandlungsgesetz,)

This legislation has been interpreted in the light of the German constitutional guarantee that states: 

‘Religious societies shall regulate and administer their affairs independently within the limits of the law that applies to all. They shall confer their offices without the involvement of central government or local authorities.’ (Grundgesetz Article 140).

The German courts have consistently taken the approach that, in the light of this constitutional guarantee of self-determination, religious institutions may decide in accordance with their own faith-defined self-perception what is necessary to fulfil their religious mission. This means that religious employers may themselves definitively determine (subject only to plausibility review by the courts) what constitutes acting ‘in good faith and with loyalty to the ethos of the organisation’ within the meaning of Article 4(2) of the Directive.

The litigation between JQ and IR spent several years moving between various levels of the German court system (including a hearing before the Bundesverfassungsgericht (Federal Constitutional Court) which did not refer the matter to the Court of Justice).

When it came before the Bundesarbeitsgericht (Federal Labour Court) for the second time, the court decided to make a reference to the Court of Justice under Article 267 in order to ascertain whether the broad scope granted by German law to religious employers to determine the scope of the duty of employees to be loyal to the employers’ ethos was compatible the provisions of Article 4(2) of the Directive. In particular, the national court wanted to know whether a religious employer of a particular faith was entitled to apply a more stringent duty of loyalty in respect of employees who are also members of the faith than is applied to those who are of a different religion or of no religion.

The Court found that German law provided excessive scope to religious employers in this regard. This was not surprising as the Court of Justice had recently come to the a similar conclusion in the Egenberger case where a non-religious woman had been denied a job in a Protestant foundation.

In that case, the Court of Justice ruled that a decision by a religious employer to discriminate against an individual on grounds of their religion must satisfy a proportionality test. This test included an obligation on the religious employer to show that discriminating on grounds of religion was necessary because of the importance of the job in question for ‘the manifestation of the religious body’s ethos or its right of autonomy’.

In addition, the Egenberger ruling stressed that the right to effective judicial protection under Article 47 of the Charter of Fundamental Rights meant that an employee claiming to be the victim of discrimination must be able to contest whether such discrimination complied with the tests of genuineness, necessity and justification before national courts. Compliance with such tests could not be decided definitively by the religious body itself, something which goes against the German approach of allowing the self-perception of the religious institution to determine this issue, subject only to plausibility review by the courts.

The Egenberger ruling therefore made it clear that discriminatory decisions by employers must be subject to objective justification, including compliance with a proportionality test (rather than the self-perception of the religious body) as well as being contestable before an independent court.

The key question in JQ v IR was whether the principles outlined in Egenberger by the Court of Justice for dealing with cases of less favourable treatment on grounds of religion in general, would also apply when the less favourable treatment related to the imposition of an obligation of loyalty towards the ethos of the employer. Unsurprisingly, the Court held that the same principles apply and held that where an employee has been accorded less favourable treatment on account of his religion (in this case by requiring greater loyalty to the employer’s Catholic ethos from Catholic employees), such discrimination must comply with a proportionality test and that compliance must be capable of being assessed by an independent court, not the religious body itself. It noted that the ability of employers to impose a duty ‘to act in good faith and with loyalty to organisation’s ethos’, provided by Article 4(2) is subject to the proviso that this duty can be imposed only ‘provided that [the Directive’s] provisions are otherwise complied with’.

This means that the lawfulness of a difference in treatment depends not on the self-perception of the employer but:

‘on the objectively verifiable existence of a direct link between the occupational requirement imposed by the employer and the activity concerned. Such a link may arise either as a result of the nature of the activity, for example where it involves taking part in the determination of the ethos of the church or organisation in question or contributing to its evangelising mission, or of the circumstances in which the activity is to be carried out, for instance, where it is necessary to ensure that the church or organisation is presented in a credible fashion to the outside world’

In addition the difference in treatment must be shown to be genuine, legitimate and justified. In this regard, the Court reiterated its ruling in Egenberger that:

‘”genuine” means that professing the religion or belief on which the ethos of the church or organisation is founded must be necessary because of the importance of the occupational activity in question for the promotion of that ethos or the exercise by the church or organisation of its right of autonomy, as recognised by Article 17 TFEU and Article 10 of the Charter’

‘”legitimate” shows that the EU legislature intended to ensure that the requirement of professing the religion or belief on which the ethos of the church or organisation is founded is not used to pursue an aim that has no connection with that ethos or with the exercise by the church or organisation of its right of autonomy’

‘”justified” implies not only that a national court can review whether the criteria laid down in Article 4(2) of Directive 2000/78 are being complied with, but also that the church or organisation imposing the occupational requirement is obliged to show, in the light of the factual circumstances of the individual case, that the alleged risk of undermining its ethos or its right of autonomy is probable and substantial, so that the imposition of such a requirement is necessary’.

This means that the imposition of a greater duty of loyalty on Catholic employees only be justified if ‘bearing in mind the nature of the occupational activities concerned or the context in which they are carried out, the religion or belief is a genuine, legitimate and justified occupational requirement in the light of that ethos’.

While it acknowledged that it was ultimately for the national court to reach a conclusion on the facts, the Court of Justice gave a strong steer to the national court noting that:

‘Adherence to that notion of marriage does not appear to be necessary for the promotion of IR’s ethos, bearing in mind the occupational activities carried out by JQ, namely the provision of medical advice and care in a hospital setting and the management of the internal medicine department which he headed. Therefore, it does not appear to be a genuine requirement of that occupational activity within the meaning of the first subparagraph of Article 4(2) of Directive 2000/78’

And that:

‘that positions of medical responsibility entailing managerial duties, similar to that occupied by JQ, were entrusted to IR employees who were not of the Catholic faith and, consequently, not subject to the same requirement to act in good faith and with loyalty to IR’s ethos’

The Court also confirmed the position it took in Egenberger that the acknowledgement in Article 17 of the Lisbon Treaty that the Union respects the status of churches and religious associations under national law merely ‘expresses the neutrality of the European Union towards the organisation by the Member States of their relations with churches and religious associations and communities, that article is not such as to exempt compliance with the criteria set out in Article 4(2) of Directive 2000/78 from effective judicial review’.

Giving Effect to the Ruling: Indirect Effect and Mangold Principles

The national court also asked a number of questions in relation to its ability to give effect to the Directive as interpreted by the Court of Justice. When making its reference, the national court indicated that it suspected, correctly as it turned out, that German law was inconsistent with the Directive. As this was a dispute between two private parties, vertical direct effect of the Directive per se was not available. In its ruling the Court of Justice took the opportunity to remind the national court that the obligation of consistent interpretation set out in Marleasing and subsequent cases, included an obligation to adapt established national case-law.

Interestingly, the Court ruled that even if it was not possible to interpret German law consistently with the directive, the national court could still give effect to Directive 2000/78 in this case by disapplying the relevant national law. In doing so it relied, as it had in Egenberger, on the controversial line of cases arising from the Mangold ruling in which the Court held that Directive 2000/78 merely codified a pre-existing EU legal obligation to respect the general principle of equal treatment which was not dependent on any implementing measures by the Member State. This obligation applied notwithstanding the fact that JQ had been fired before the Charter of Fundamental Rights came into force because the principle of equal treatment was already binding as it was a general principle of law arising from the common constitutional traditions of the Member State.

Significance

With four major rulings in 18 months, the outlines of the approach of the Court of Justice to the religion-related provisions of Directive 2000/78 are now reasonably clear, giving us a clearer picture of the approach of the Union to religion’s role in the legal system more generally.

First, it is clear that the Court is committed to an approach that involves balancing of clashing rights through the framework of proportionality. Under EU law there is a sliding scale of religious autonomy with decreasing autonomy for religious employers the more distant a role is from the core religious functions. Furthermore, any discriminatory decisions must be capable of being challenged in a meaningful way before the courts. Religious bodies cannot determine for themselves the degree of exemption from anti-discrimination rules necessary to protect their ethos.

This is in contrast to the approach of the US Supreme Court which has adopted an approach based on the idea of a ‘ministerial exemption’ under which the state has no right to assess decisions of religious bodies in relation to roles that have a religious element and indeed, to the approach of the European Court of Human Rights which upheld, in cases such as Fernández-Martínez v Spain, an arrangement under which the automatic termination of the contract of a religion teacher in a public school on foot of decision of a local bishop to withdraw his endorsement of that teacher was subject to very limited review. The commitment of the Court to proportionality as a means to resolves clashes in this area means that it is likely that, when it is faced with a case of the non-discriminatory application of an ethos-loyalty obligation to an employee, it will insist that such an obligation be limited in order to ensure that it does not have a disproportionate impact on other fundamental rights such as the right to privacy of freedom of expression.

Second, the Court is keen to limit its tolerance of policies that have potentially discriminatory implications on grounds of religion, to instances where the policy is consistently applied. Thus, in Achbita and Bougnaoui it stressed that neutrality policies would be considered indirectly rather than directly discriminatory only if consistently applied to all statement of all forms of religious, philosophical and political belief. Similarly, in JQ v IR it found fault in the fact that the loyalty obligation would not have been applied to a non-Catholic in JQ’s role. The Court has also made it clear that, as in Bougnaoui, in cases where a policy is shown to be directly discriminatory it will not give a wide interpretation to the concept of a ‘genuine and determining occupational requirement’. This approach also involves a commitment to treat religious and non-religious beliefs equally, something that sets EU law apart from approaches in many other areas of the world.

Finally, the Court regards the Treaty commitment in Article 17 to respect the status of religious bodies in national law not as a mandate to exempt the legal privileges of religious bodies from review for their compliance with EU legal norms but as a statement of the EU’s neutrality in relation to the different ways in which Member States organise their relations with churches and religious bodies.

Finally, for EU law more generally, recent religion cases have underscored the Court’s determination to persevere with its controversial approach of regarding Directive 2000/78 as doing no more than giving specific expression to an already self-executing general principle of non-discrimination in EU law.

Barnard and Peers: chapter 9, chapter 20

Photo credit: Lifenews.com